Theory of Mind and Self-Concept

Abstract

Preschool participants from Korea (N = 56, mean age = 5 years, 4 months) and the United States (N = 56, mean age = 5 years, 1 month) responded to tasks assessing self-concept and understanding of false belief. Korean children scored higher on the self-dimension of Traditionalism and lower on the dimension of Social Potency than did American children, a finding in accord with the hypothesis that the Korean culture fosters a relatively interdependent conception of the self. Korean children also outperformed American children on the measure of false belief. In contrast to previous research with Western samples, the nature of the false belief problem (inanimate object, voluntary person movement, involuntary person movement) did not significantly affect performance in either culture. Finally, variations in self-concept showed some but limited relations to false belief performance; in general, relatively interdependent self-responses were predictive of good false belief performance.

Keywords

theory of mind false belief self-concept cross-cultural

Research under the heading of theory of mind is directed to children’s understanding of the mental world—what they know or believe about mental phenomena such as thoughts, beliefs, and desires. Such research has been a popular enterprise for close to two decades now, and it has demonstrated that important advances in children’s understanding of mental phenomena typically occur between the ages of 3 and 5. Prominent among these advances is mastery of false belief: the realization that people may hold beliefs that are not true.

A common assumption in theory-of-mind research is that developments such as false belief are universal achievements that should appear in any cultural setting. Such developments have in fact now been demonstrated in a range of settings, including some cultures that are quite distinct from the Western industrialized societies with which research began; examples of the latter include Cameroon (Avis & Harris, 1991), Papua New Guinea (Vinden, 1999), and Samoa (Callaghan et al., 2005). Nevertheless, the empirical literature on theory of mind remains largely a Western literature, with the great majority of studies coming from the United States, Canada, and Western Europe. In this report, we provide a comparison of false belief performance in a Korean sample and an American sample. We also compare aspects of self-concept in the two samples, and we explore possible relations between the self-measure and different forms of false belief understanding.

Two meta-analyses by Wellman and colleagues summarize most of the previous false belief research with Asian samples (Liu, Wellman, Tardif, & Sabbagh, 2008; Wellman, Cross, & Watson, 2001). The Wellman et al. (2001) review provides a summary of false belief studies in general through 2000, with culture as one of the “moderator variables” analyzed. At that point two Asian countries, Japan and Korea, had a sufficient research literature to merit inclusion in the meta-analysis. Wellman et al. concluded that Korean children appeared to show the same timing of mastery as that typical for Western samples, but that Japanese children lagged somewhat behind.

The literature on Chinese children grew rapidly in the years following the 2001 publication, and this literature was the focus of the second meta-analysis. Liu et al. (2008) reported that Chinese children demonstrated the same trajectory of mastery as Western samples—that is, they showed the same predictable progression from below-chance to above-chance performance with age, and the effects of various task manipulations were equivalent in the two groups. Children from mainland China were found to develop at the same pace as U.S. samples, reaching above-chance performance at approximately 50 months, but more slowly, by about a year, than Canadian samples. Children from Hong Kong lagged behind both Western groups, reaching above-chance performance only at about 64 months (Liu et al., 2008).

Several recent reports add to the comparative picture. Using a battery of theory-of-mind tasks, both Sabbagh, Fu, Carlson, Moses, and Lee (2006) and Wellman, Fang, Liu, Zhu, and Liu (2006) confirmed the equivalent performance of Chinese mainland and U.S. samples. Although no within-study comparison was provided, two studies by Cheung and associates (Cheung, 2006; Cheung et al., 2004) reported theory-of-mind performance by Chinese preschoolers that was comparable to that usually found in Western samples. In contrast, a series of studies by Naito and Koyama (2006) provided further evidence of the relatively slow development of Japanese children. Naito and Koyama found that Japanese children lagged about a year to a year and a half behind the trajectory of mastery shown in the most directly comparable Western research. Similar results have emerged in other recent research in Japan (Naito & Seki, 2009; Ohtsubo, 2007; Toyama, 2007).

The Naito and Koyama (2006) research also included an examination of different forms of the unexpected transfer false belief task. In the typical form of the task, an inanimate object is moved without the protagonist’s knowledge. Naito and Koyama compared performance on the typical task with performance on two versions of the task in which a person rather than an object moved, in one case voluntarily and in one case in response to a command from another. The point of the comparison was that several studies with Western samples (Rai & Mitchell, 2004; Symons & Clark, 2000; Symons, McLaughlin, Moore, & Morine, 1997) had reported that children found the voluntary-movement trial especially difficult. The explanation offered for this finding was that voluntary movement by an animate object introduces a second set of mental states for children to consider—not only the belief of the target for the false belief question but also the thoughts and intentions of the self-moving agent. Even though only the former belief is relevant in this context, children who are in the process of mastering false belief apparently find the inclusion of additional mental states confusing and thus perform more poorly. Nguyen and Frye (1999) reported a result consistent with this explanation: Children performed more poorly in judging a false belief regarding a play partner’s intended activity than they did on a standard false belief trial.

Naito and Koyama (2006) grounded their task comparison in a commonly held conception of Japanese culture specifically and of Asian culture more generally (Markus & Kitayama, 1991; Nisbett, 2003; Sampson, 1988). The conception is that Asian cultures are characterized by an emphasis on interpersonal relations and interdependency, an emphasis that fosters a person- oriented, collectivist conception of the self. In contrast, Western cultures are said to emphasize independence and autonomy and thus the development of a more autonomous, individualistic self-concept. With respect to the false belief task, Naito and Koyama argued that Japanese children, with their strong focus on social relations and social rules, might focus less on what the characters were thinking and feeling and more on the situational determinants of how one should behave—specifically, that a child seeking to reunite with a caregiver (as on the Symons et al., 1997, person trials) should go to that person’s actual location. If so, they would be expected to perform relatively poorly on the false belief task, given that it is the character’s belief rather than the reality that should govern response on this task. But they would not be expected to have special difficulty with the additional mental states introduced in the voluntary-movement case, given that their focus was situational rather than mentalistic.

Their data supported both predictions. As already noted, Japanese children performed more poorly than the typical level of performance in Western samples. In contrast to the Western participants in earlier studies, however, they showed no difference in response to the three sorts of false belief trials. The results thus suggested not only a quantitative difference (slower development) but also a qualitative difference between Japanese children’s and Western children’s reasoning about beliefs. In Naito and Koyama’s (2006) analysis, both findings stemmed from the same source: the relatively situational as opposed to mentalistic orientation of the Japanese children.

One goal of our study was to extend the comparison of different forms of false belief to a Korean sample. As noted, Wellman et al. (2001) concluded that Korean children performed comparably to Western children; this conclusion, however, was based on a single multistudy report that did not include a Western comparison sample (Ghim, 1997). Since their review, one comparative study has appeared: Oh and Lewis (2008) reported that Korean preschoolers performed equivalently on standard false belief measures to a sample of English preschoolers from Lancaster, United Kingdom. There has been one false belief study with Korean American children, and it found that such children outperformed their Anglo counterparts (Vinden, 2001). Our study is the first to provide a within-study comparison of Korean and American children. It is also the first to examine the three different forms of false belief problem in a Korean sample.

A second goal was to provide a further test of the voluntary-movement variable. As noted, Naito and Koyama (2006) failed to replicate the effect from the original research, a finding that they attributed to an Eastern-Western cultural difference. Their procedure, however, was in several respects deliberately different from that in the original studies by Symons and colleagues (Symons & Clark, 2000; Symons et al., 1997). The Symons studies included a question about the target’s emotion prior to the question about belief, the belief question was worded in terms of “do” rather than the more common “look” or “think,” and the person trials included two movements rather than the typical single movement. The Naito and Koyama procedure, in contrast, was modeled more closely after the standard object location task. Our procedure was as well, thus providing a test of the generality of the voluntary-movement effect in Western samples, as well as a further test of a possible cultural difference between Asian and Western children.

A third goal of the study was to assess the hypothesized cultural difference in self-concept more directly. Although there is a fairly large empirical literature in support of the independent-interdependent distinction, the great majority of studies are with adults. Most of the evidence for comparable effects in childhood is indirect. Two related lines of research are most often cited. First, a number of studies have demonstrated on-the-average differences between Eastern and Western cultures in how mothers talk to young children about past events; Western mothers engage in more such talk than do Asian mothers, and they are more likely to focus the talk on experiences and attributes of the child (Leichtman, Wang, & Pillemer, 2003). Second, comparable effects have been demonstrated in how children themselves talk about past events; Western children talk about the past more often, their memory talk includes a higher proportion of references to the self, and as adults they report earlier autobiographical memories (Han, Leichtman, & Wang, 1998; Mullen, 1994). Korea is among the countries for which such Western-Eastern differences have been demonstrated at both the child and adult levels (Choi, 1992; Mullen & Yi, 1995).

To our knowledge, the only direct examination of self-concept in Korean children is a study by Lee, Super, and Harkness (2003). The measures were culturally adapted versions of two scales developed by Harter: the Self-Perception Profile for Children and the Pictorial Scale of Perceived Competence and Social Acceptance for Young Children (Harter, 1985; Harter & Pike, 1984). The adaptations for the Korean sample included the addition of items intended to assess social competencies of a collectivist sort (e.g., “greet elderly neighbors,” “help a younger child who falls”). Although admittedly preliminary, the findings provided some evidence that the collectivist dimension did in fact constitute an independent dimension of self-evaluation for Korean children.

Self-esteem measures such as the Harter scales focus on the evaluative component of the self, what is commonly labeled self-esteem or self-worth (Harter, 2006). Our interest was more in the cognitive component—what children understand or believe about their attributes across different dimensions of their lives. We focused on the cognitive aspect rather than the evaluative aspect of the self for two reasons. First, as our earlier review indicated, it is the cognitive aspect of the self that is stressed with regard to hypothesized cultural difference in self-conceptions (Markus & Kitayama, 1991). Second, we hypothesized that the cognitive aspect of the self would be most likely to affect beliefs about others and thus theory-of-mind performance. The measure we selected for the self-assessment was the Eder Self-View Questionnaire (Eder, 1990). The Self-View Questionnaire is for children between 4 and 8, and it uses a simple, two-choice response format to assess self-concepts across 10 dimensions: Achievement, Aggression, Alienation, Control, Harm-Avoidance, Social Closeness, Social Potency, Stress Reaction, Traditionalism, and Well-Being.

To our knowledge, the only cross-cultural application of the Eder measure is a study by Hayward (2004). Hayward compared samples of 5- and 6-year-olds from America and Japan. Of the 11 dimensions assessed (the 10 original Eder dimensions and a new scale, Collective Identity, added for the study), three showed significant differences: Japanese children scored higher on Collective Identity, Harm-Avoidance, and Social Closeness than did American children. In general, then, the results were in accord with expectations based on the interdependent-independent distinction.

A final goal of the study was to explore possible relations between self-concept and false belief mastery. Naito and Koyama (2006) suggested that the different trajectories of mastery shown by Japanese children and Western children reflected differences in self-development that impacted response to the false belief problems. Their study, however, did not include a measure of self-concept; thus, there was no possibility of demonstrating within-child relations between self-conceptions and false belief performance. To date, most examinations of possible relations between self-development and theory of mind have focused on the regulative aspect of the self as expressed on measures of so-called executive function (Hughes, 2002). A study by Cahill and associates does report some associations between self-evaluations and theory of mind, although the causal directions stressed is from theory of mind to self rather than the reverse (Cahill, Deater-Deckard, Pike, & Hughes, 2007). Our study is the first to include a measure of self-cognitions along with false belief, thus allowing a determination of whether variations in self-concept relate to false belief performance both for the sample as a whole and within each culture separately.

Time constraints prevented administration of the entire Eder Self-View measure. We included the four scales that seemed to us most relevant to the interdependent-independent distinction: Traditionalism, Achievement, Social Closeness, and Social Potency.¹ (See Table 1 for a description of the scales.) The scales were selected such that there were two (Traditionalism and Social Closeness) on which Korean children were expected to score higher and two (Achievement and Social Potency) on which American children were expected to score higher. The selections and predictions were grounded in the literature on interdependent versus independent conceptions of self (e.g., Oyserman, Coon, & Kemmelmeier, 2002) and were also based on the experiences of the first author, a native of Korea who had lived there until shortly prior to the inception of the research. In addition to the assessment of self-concept, each child responded to three false belief tasks: a standard task on which an object was moved and two tasks on which a person moved, in one case voluntarily and in one case in response to a command. As noted, four issues were of interest.

Table 1.

Examples of Items on the Self-View Questionnaire

Dimension	Definition^a	High Statement	Low Statement
Achievement	Works hard; enjoys demanding activities; is a perfectionist	I like hard work. I care about doing a really good job at everything I do.	I don’t like hard work.
			I don’t care about doing a really good job at everything I do.
Social Closeness	Friendship dimension: seeks intimacy; loves people; helps	I like to play with friends. When I am sad, I go find someone to play with.	I like to play by myself.
			When I am sad, I go play in my room by myself.
Social Potency	Leadership dimension: likes to stand out, influence people, be the center of attention	I pick the game to play. I like to show things at “Show and Tell” at school.	I don’t pick the game to play.
			I don’t like to show things at “Show and Tell” at school.
Traditionalism	Authority dimension: cares about manners, being polite.	I never do things I’m not supposed to do. I usually do what Mommy or my teacher says.	I sometimes do things I’m not supposed to do. Sometimes I don’t do what Mommy or my teacher says.

Definitions are from Eder (1990).

One question concerned possible differences in self-concept between Korean children and American children. In line with the independent-interdependent distinction, we expected Korean children to score higher on Social Closeness and Traditionalism and lower on Achievement and Social Potency than American children.

A second question concerned possible differences in false belief mastery between the two samples. Given the inconsistencies in prior Asian-Western comparisons, we regarded the false belief cultural comparison as an open question.

A third question concerned possible differences in difficulty among the false belief tasks. We expected American children to have special difficulty with the Person Internal task, in accord with the previous examinations of this issue. Naito and Koyama (2006) reported that Japanese children had no special difficulty with this problem, presumably because of their situational rather than mentalistic focus. Assuming a similar situational focus in our sample, we expected Korean children to perform equivalently on the various problems.

The final issue concerned possible relations between self-concept and false belief mastery. Based on the cultural differences in false belief performance reported by Naito and Koyama (2006), our expectation was that relatively interdependent self-conceptions (high on Traditionalism and Social Closeness, low on Achievement and Social Potency) would be predictive of a situational rather than mentalistic orientation and thus of relatively poor false belief performance.

Method

Participants

The participants included 56 Korean preschoolers and 56 American preschoolers. The Korean sample consisted of 19 four-year-olds (9 boys and 10 girls, mean age = 4 years, 8 months, range = 4-1 to 4-11) and 37 five-year-olds (21 boys and 16 girls, mean age = 5 years, 8 months, range = 5-0 to 6-0). The mean age for the sample as a whole was 5 years, 4 months. The American sample included 20 four-year-olds (8 boys and 12 girls, mean age = 4-7, range = 4-0 to 4-11) and 36 five-year-olds (23 boys and 13 girls, mean age = 5 years, 5 months, range = 5-0 to 6-0). The mean age for the sample as a whole was 5 years, 1 month. The American sample was primarily Caucasian, with the exception of 2 African American children and 1 Hispanic American child. The sample did not include any Asian or Asian American children.

The Korean sample was drawn from government-supported preschools and public kindergarten classes in Seoul and in Bucheon, a city near Seoul. Socioeconomic status (SES) disparities across different regions of Korea are marked, and both western Seoul and Bucheon comprise primarily lower-middle-class populations (Bae & Lee, 2008). The American sample was drawn from a mixture of private and university-affiliated daycare centers in a midsize university community in the Southeast. Although we did not collect systematic data on rate of participation among prospective participants, the participation rate was higher in Korea (approximately 80% to 90%) than in America.

Procedure

Each child participated in a single experimental session administered by two college-aged female testers. The design was within-subject, with each child responding to the three false belief tasks and the self-measure. The false belief measures were given first, followed by the self-measure.

The testing in Korea was conducted in Korean. The first author (a bilingual speaker of Korean and English with several years of experience as a preschool teacher in Korea) translated the scripts from English to Korean. Two preschool teachers assisted with the translation; they judged the overall appropriateness of the wording and identified any words or phrases that might be subject to misunderstanding. For the Korean testing, the children in the false belief stories were given Korean names and the hair colors were changed to brown; similarly, the dolls in the self-test were given Korean names. Otherwise, the stimuli and procedure were identical in the two countries.

False belief

There were three false belief tasks, all of the unexpected transfer form. On each, the target for the protagonist’s search was moved without the protagonist’s knowledge, and the question concerned where the protagonist would look for the target. The tasks differed with respect to the nature of the target. On the Object task a physical object was moved, on the Person Internal task a person moved as a result of his or her own volition, and on the Person External task a person moved in response to a command from another.

The order of the three types of tasks was balanced across children. In addition, two versions of each type of task were used, with half the children receiving one version and half receiving the other. The purpose of the two versions was to ensure that a particular comparison (in particular, Person Internal vs. Person External) was not confounded with the specific scenarios used. The appendix provides an example of one of the two versions. In the alternative version the two Person tasks were reversed; that is, through a slight change in wording the movie theater scenario became the Person Internal task and the schoolyard scenario became the Person External task. On the Object task, the kite was replaced by a book, which one character moved in the other character’s absence.

The stories were presented in story book form, with each story accompanied by three 8 × 11 inch colored illustrations. The first illustration depicted the starting-point situation (e.g., Jenny and Billy playing with kites), the second depicted the original location at the time of the protagonist’s departure (e.g., Jenny placing the kite by the rock as her mother calls to her from the house), and the third depicted the final location (e.g., the kite in the tree).

As the appendix indicates, each story ended with the experimenter reminding the child of the initial situation prior to the protagonist’s departure (e.g., “When Jenny last saw her kite it was by the rock.”). This was followed by a control question directed to the current reality (e.g., “Where is Jenny’s kite now?”). If the child failed to answer the reality question correctly, the critical part of the scenario was repeated and the question was asked again. If the child failed to answer correctly after three attempts, the task was omitted. The test question followed. It was asked originally in open-ended form (e.g., “Where will Jenny look for her kite?”); if the child failed to respond, a closed-choice follow-up was used. Responses, whether to the open-ended or closed-choice version, were scored as either pass or fail.

Self-concept

The Self-View Questionnaire (Eder, 1990) is a self-report, forced-choice measure in which the child indicates which of two opposing descriptive statements is more true of himself or herself. The statements are presented in the context of a puppet show, in which two puppets or dolls provide the contrasting descriptions. In our study the stimuli were four 50 cm tall cloth dolls. Boy dolls were used for boys, and girl dolls were used for girls.

The same tester who had administered the false belief trials presented the Self-View task. After introducing the two dolls, she explained: “These two dolls are writing a story about kids your age. They want to learn all about you. They are going to tell you about themselves and then you tell them about yourself.”

Two practice trials followed. On the first, one doll said, “I go to school,” and the other doll said, “I don’t go to school,” after which the tester asked, “What about you?” On the second practice trial, one doll said, “I like ice cream,” and the other said, “I don’t like ice cream,” followed by the “What about you?” question. Prompting if necessary and feedback and encouragement were given on both practice trials.

The full Self-View Questionnaire includes 50 items distributed across the 10 dimensions listed earlier. As noted, our assessment focused on the four dimensions that seemed to us most relevant to the cultural comparison we wished to make: Achievement, Social Closeness, Social Potency, and Traditionalism. Thus we administered 20 items, 5 from each of these 4 dimensions. Samples of the items are given in Table 1.

A single, semirandomized order of presentation was used, with the constraint that no more than two items from the same dimension appear in succession. The positive statement was given first on half the trials, and the negative statement was first on the other half. In addition, the doll on the right spoke first on half the trials, and the doll on the left spoke first on the other half.

One item from the Social Potency scale (“I’m the leader in ‘Follow the leader’”) was judged not to be relevant in Korea (where there is no game similar to “Follow the leader”) and therefore was omitted in the Korean testing.

Results

False Belief

For their responses to be included on a false belief task, children needed to answer the reality control question correctly. There were 45 instances (of a possible 336) in which children initially answered the question incorrectly and required a repetition: 30 in Korea (21 instances for 4-year-olds and 9 for 5-year-olds) and 15 in America (7 instances for 4-year-olds, 8 for 5-year-olds). There were four instances in which children did not answer the question correctly even after three attempts: three in Korea (two 4-year-olds, one 5-year-old) and one (a 5-year-old) in America. These trials were omitted in the analyses that follow.

Table 2 shows the proportion of correct responses on each of the three forms of false belief task both within each culture and for the sample as a whole. These data were analyzed via generalized estimating equation (GEE) analyses (Liang & Zeger, 1986; Zeger & Liang, 1986) appropriate for a repeated-measures design with dichotomous dependent variables. The predictors were culture, age, and task, as well as the possible two-way interactions among the three predictors.

Table 2.

Proportion of Correct Responses on the Three False Belief Trials

	Object	Person External	Person Internal	Total
Korea	49	63	56	56
United States	34	27	31	31
Total	41	45	43

The analyses yielded one significant main effect, one marginally significant main effect, and one significant interaction. Korean children outperformed American children, χ²(1) = 9.40, p = .002. This finding was qualified, however, by a significant interaction of culture and task, χ²(2) = 6.23, p = .044. Follow-up tests indicated that Korean children performed better than American children on both the Person Internal task, χ²(1) = 6.65, p = .009, and the Person External task, χ²(1) = 13.52, p =.001; the difference on the Object task was not significant. Finally, there was a marginally significant main effect of age, χ²(1) = 3.66, p = .056, with 5-year-olds (M = .49, SD =.50) outperforming 4-year-olds (M = .32, SD = .47).

Self-Concept

Children earned 1 point for each endorsement of a High statement and 0 points for each endorsement of a Low statement. As noted, one item on the Social Potency scale was omitted for the Korean children. In the analyses that follow, the scores on that scale have been multiplied by 1.25 to make them comparable with the other scales.

Table 3 presents the mean self scores on the four dimensions. These scores were examined in a 2 (Age) × 2 (Culture) × 4 (Self-Domain) ANOVA. This analysis yielded a significant main effect of domain, F(3, 324) = 13.25, p < .001, and a significant interaction of domain and culture, F(3, 324) = 8.59, p < .001. Follow-up tests of the main effect (Bonferroni post hoc pairwise comparisons) indicated that scores were significantly higher on Social Closeness than on the other three dimensions, which did not differ from each other. Follow-ups of the interaction indicated that Korean children scored higher than American children on Traditionalism and American children scored higher than Korean children on Social Potency. There were no differences between cultures on the other two dimensions.

Table 3.

Means and Standard Deviations for the Four Self-Concept Dimensions

	Social Closeness		Traditionalism		Social Potency		Achievement
	M	SD	M	SD	M	SD	M	SD
Korea	4.20	1.02	3.96	1.11	2.92	1.24	3.18	1.05
United States	4.29	0.89	3.29	1.28	3.72	1.14	3.50	1.25
Total	4.25	0.95	3.63	1.24	3.32	1.26	3.34	1.16

The possible range of scores is from 0 to 5. Children from Korea received only four Social Potency trials; their scores have been adjusted to place them on a 5-point scale.

Possible relations among the four self-dimensions were examined through Pearson correlations, calculated both for the sample as a whole and separately within culture. These analyses revealed a significant correlation between Achievement and Social Potency for the sample as a whole, r(112) = .21, p < .05; the correlation disappeared, however, when age was partialed out. There were no significant correlations within either the Korean or the American sample. The independence of the various dimensions is in accord with the original Eder (1990) study, which reported sporadic correlations between dimensions, but none for the dimensions or the age group examined here.

Relations Between Self-Concept and False Belief

Possible relations between the self-scores and false belief performance were examined in three sets of logistic regression analyses, first for the sample as a whole and then separately for the Korean and the American children. The predictors were age and each of the four self-scores. Because these analyses overlap with analyses reported earlier, we concentrate here on the findings specific to the self-measures.

The analysis for the total sample produced two main effects of the self-concept scores. The analysis on the Person External task, χ²(3, N = 111) = 13.06, p = .005, revealed a main effect of Social Potency (β = –.48, Wald = 4.63, p = .031), and that on the Person Internal task, χ²(3, N = 109) = 13.26, p = .004, revealed a significant effect of Achievement (β = –.36, Wald = 3.99, p = .046). In both cases, the relation was a negative one: Better false belief performance was associated with lower scores on the self-concept dimension. The main effect of Social Potency on Person External was qualified by a significant interaction with age (β = –.86, Wald = 4.21, p = .04), reflecting the fact that the effect was contributed largely by the older children. The analysis of the Object task, χ²(3, N = 111) = 8.13, p = .043, revealed a similar interaction, in this case of Social Potency and age (β = –.86, Wald = 4.13, p = .042): For younger children, higher scores on Social Potency were associated with better false belief performance, whereas for older children the relation was the reverse.

The final pair of analyses examined whether the effects found for the sample as a whole held within each culture separately. These analyses, it should be noted, entail a loss of power, given both the reduction in sample size and the lesser variability in response within than across cultures. With this qualification, the answer to the question was no: The number of significant effects from the two regressions did not exceed what would be expected by chance. Thus, there was no evidence for links between self-conceptions and false belief within culture.

Discussion

Studies that compare false belief mastery in Asian and Western samples have produced all three possible outcomes: equivalent performance, better performance by Western children, and better performance by Asian children. Our study is another entry in the last of these categories. The Korean children in our study outperformed the American children on both the false belief battery considered as a whole and the two Person trials considered separately.

It is not possible, of course, to draw general conclusions about differences between cultures from a single study with limited measures and limited samples. A recent report by Wellman et al. (2006) makes the point with respect to American-Chinese comparisons:

Yet children from the United States and China, even if carefully sampled for comparable ages and socioeconomic status, differ widely (e.g., in the languages they acquire, milestones of language acquisition, family experiences, the nature of schools and preschools they attend, timing of entry into schools). Mean group differences on some task or achievement may always be due to those uncontrollable confounding factors (rather than focal conceptions or skills). (p. 1077)

In the present instance, various factors may have affected the comparison, perhaps most obviously the specific samples that were studied. As noted, the Korean sample came from more urban and suburban environments than did the American sample; they were probably (although we did not collect systematic data) from lower-income families, and they may have represented a less select subset of their cohort, given the higher participation rate in Korea. There also, of course, are differences in children’s educational experiences between the two settings (Kwon, 2002, 2003), although such differences are probably best interpreted not as confounds but as intrinsic to the cultural comparison. In any case, it is worth noting that most of the disparities we are able to identify would seem to favor the American children; in particular, a number of studies (e.g., Holmes, Black, & Miller, 1996; Shatz, Diesendruck, Martinez-Beck, & Akar, 2003) have documented slower theory-of-mind development for lower-income children. Yet it was the American children who performed more poorly on false belief. At the least, then, our results, in conjunction with previous studies of Korean and Korean American children (Ghim, 1997; Oh & Lewis, 2008; Vinden, 2001), suggest that Korean children, in contrast to Japanese children, do not experience any lag relative to Western samples in their mastery of false belief.

We agree with Wellman et al. (2006) that differences in rate of development are less interesting, in part because they are less interpretable given the difficulty of matching samples, than are possible qualitative differences in how children of different cultures approach false belief problems. As noted, Naito and Koyama (2006) argued that their results showed such a qualitative difference, in that Japanese children’s performance, in contrast to the performance of Western children, was not disrupted by the additional mental states introduced on the Person Internal task. The Korean children in our study also showed no disruption on this task. Indeed, they performed better on the two Person tasks than on the Object task (although the difference fell short of significance), and they significantly outperformed the American children only on the two Person tasks.

The relatively good performance of the Korean children on the Person tasks suggests a different contribution from the self-system than that hypothesized by Naito and Koyama (2006). They argued that the interdependent nature of Japanese children’s self-concepts led to a focus on the situational determinants of proper behavior in social situations, with a resulting relative neglect of mental states. Our results suggest that the interdependent, person-oriented nature of Korean children’s self-concepts may make them especially attuned to the thoughts and wishes of others, thus leading to better performance on false belief tasks in general and on tasks with social referents in particular. Such an outcome is compatible with the recently formulated Functional-Multilinear Socialization theory of theory of mind of Lucariello and colleagues, which argues explicitly that upbringing in a collectivist culture should foster understanding of the beliefs of others (Lucariello, 2004; Lucariello, Le Donne, Durand, & Yarnell, 2006).

The preceding argument assumes that the Korean children did in fact possess a more interdependent self-system than the American children. The results from the Self-View Questionnaire provide some support for this conclusion. As expected, Korean children were higher in Traditionalism and lower in Social Potency than American children, and the mean difference in Achievement was also in the expected direction. The results from the Lee et al. (2003) study summarized earlier provide further evidence for a collectivist dimension in Korean children’s self-concepts.

Having emphasized the supposed interdependent, collectivist nature of Asian societies, we should add two points made by many authors who write about the issue (e.g., Hermans & Kempen, 1998; Mascolo & Li, 2004; Oyserman et al., 2002; Raeff, 2006). The first point is that Asian cultures, although similar in some respects, are not homogeneous, and an exclusive focus on their supposed interdependency and collectivism may obscure both important aspects of each culture and important differences among cultures. There is evidence, for example, that both Korea and Japan are less collectivist and more individualist than is China (at least at the adult level—Oyserman et al., 2002), as well as evidence for changes in these dimensions over time (Cha, 1994). And of course, the varied conclusions from theory-of-mind research with Asian samples, including the present results, suggest that children in different Asian countries encounter important differences in socialization experiences.

The second point concerns prediction of individual development. Whatever the on-the-average differences may be between cultures, these patterns clearly do not apply to all children within a culture. The present study moves beyond culture-level comparisons to assess the characteristics of interest at the individual level, thus allowing an examination of the relation between the two constructs not only at a group level but also within-child. The relations that emerged between the two measures were modest, and they emerged only for the sample as a whole and not separately within culture. In general, however, it was the less individualistic, more interdependent pattern of response—lower scores on Social Potency, lower scores on Achievement—that appeared conducive to false belief mastery. Thus, the country-level comparisons and the individual correlations produced similar conclusions with regard to how aspects of the self-system may impact theory-of-mind development. We should add, however, that such correlational relations are also subject to the reverse causal interpretation, since there is evidence that theory-of-mind abilities can affect children’s self-conceptions (Cahill et al., 2007; Cutting & Dunn, 2002).

Two findings still require consideration. One concerns the discrepancy between the relatively good performance of the Korean children in the present study and the relatively poor performance of Japanese children that has now been reported in a number of studies (Naito, 2003; Naito & Koyama, 2006; Naito & Seki, 2009; Ohtsubo, 2007; Toyama, 2007). There is no way, of course, for us to resolve this issue with the present data, but we can suggest several possibilities for future study. The simplest explanation concerns the samples that have been studied. Both Korea and Japan are diverse nations with substantial variations across geographical regions (e.g., in income level, urban vs. rural residence) that may impact children’s development, and the sampling to date from both countries is too limited to be confident that general population differences have been identified. A second possibility is that differences in language contribute to differences in performance. Aspects of Cantonese have been shown to affect false belief reasoning in Chinese children (Tardiff, Wellman, & Cheung, 2004), and Naito and Seki (2009) point out ways in which the Japanese language may influence how children reason about beliefs. A third possible direction for study comes from recent work on the neural underpinnings of theory-of-mind reasoning, work that has revealed some on-the-average differences between Western participants and Asian participants in both child and adult samples (Han & Northoff, 2008; Kobayashi, Glover, & Temple, 2007). Finally, it is possible that either the degree of interdependence or the form it takes varies, at least at the childhood level, between Korea and Japan. By some measures, Western-Eastern differences in self-conceptions are more marked for Korean adults than for Japanese adults (Oyserman et al., 2002). Possible differences in childhood remain to be explored.

The second finding is an aspect of the American children’s data. It concerns response to the Person Internal task. Our results did not replicate those from previous studies of this manipulation with Western samples (Rai & Mitchell, 2004; Symons & Clark, 2000; Symons et al., 1997). Naito and Koyama (2006) noted some possible methodological problems in the original studies, and their modified version of the task also did not confirm its greater difficulty (though only in Japanese children, since they did not include a Western comparison group). As noted, our version of the task, like theirs, matched the standard false belief task more closely than the versions used in the Symons and Rai and Mitchell studies, and our results also showed no special difficulty in the voluntary-movement case. We have since applied similar procedures in a second study with Korean preschoolers and again found no difference between the person and object tasks (Ahn & Miller, 2007). These various null outcomes do not negate the previous findings, which have proved robust across several studies, but they do suggest that the voluntary-movement effect is more limited and task-specific than originally assumed.

In summary, in the present report we provide data on a number of issues that have received either limited or no attention in previous research. Our study is the first to relate self-concept to false belief performance, the first to examine the person false belief task in an Asian sample, the first to compare Korean and American children in self-concept, and the first to compare Korean and American children in false belief performance. Our results, we believe, add to the literatures on the cultural bases for both self-development and theory of mind, while also suggesting fruitful directions for future research.

Footnotes

Appendix

Acknowledgements

We are grateful to Hyeyoung Ahn and Kiyoung Chung for their help with recruitment and testing. We also thank James Algina for his advice with regard to statistics.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Ahn

Miller

S. A.

(2007, March). Connections between cognitive and self development in a preschool sample. Paper presented at the meeting of the Society for Research in Child Development, Boston.

Avis

Harris

P. L.

(1991). Belief-desire reasoning among Baka children: Evidence for a universal conception of mind. Child Development, 62, 460-467.

Bae

Lee

(2008). How to resolve socio-economic gaps existing in Seoul? (SDI Policy Report, 16). Seoul, Korea: Seoul Development Institute

Cahill

K. R.

Deater-Deckard

Pike

Hughes

(2007). Theory of mind, self-worth, and the mother-child relationship. Social Development, 16, 45-56.

Callaghan

Rochat

Lillard

Claux

M. L.

Odden

Itakura

. (2005). Synchrony in the onset of mental-state reasoning: Evidence from five cultures. Psychological Science, 16, 378-384.

Cha

(1994). Aspects of individualism and collectivism in Korea. In Kim

Triandis

H. C.

Kagitcibasi

Choi

Yoon

(Eds.), Individualism and collectivism (pp. 157-174). Thousand Oaks, CA: Sage.

Cheung

(2006). False belief and language comprehension in Cantonese-speaking children. Journal of Experimental Child Psychology, 95, 79-98.

Cheung

Hsuan-Chih

Creed

Wang

S. P.

(2004). Relative roles of general and complementation language in theory-of-mind development: Evidence from Cantonese and English. Child Development, 75, 1155-1170.

Choi

S. H.

(1992). Communicative socialization processes: Korea and Canada. In Iwasaki

Kashima

Leung

(Eds.), Innovations in cross-cultural psychology (pp. 103-122). Amsterdam: Swets & Zeitlinger.

10.

Cutting

A. L.

Dunn

(2002). The cost of understanding other people: Social cognition predicts young children’s sensitivity to criticism. Journal of Child Psychology and Psychiatry and Allied Disciplines, 43, 849-860.

11.

Eder

R. A.

(1990). Uncovering young children’s psychological selves: Individual and developmental differences. Child Development, 61, 849-863.

12.

Ghim

(1997). Development of children’s theory of mind: Children’s understanding of false belief. Korean Journal of Developmental Psychology, 10, 74-91.

13.

Han

J. J.

Leichtman

M. D.

Wang

(1998). Autobiographical memory in Korean, Chinese, and American children. Developmental Psychology, 34, 701-713.

14.

Han

Northoff

(2008). Culture-sensitive neural substrates of human cognition: A transcultural neuroimaging approach. Nature Reviews Neuroscience, 9, 646-654.

15.

Harter

(1985). The self-perception profile for children. Denver, CO: University of Denver.

16.

Harter

(2006). The self. In Damon

Lerner

R. M.

(Series Eds.) & Eisenberg

(Vol. Ed.), Handbook of child psychology. Vol. 3: Socialization and personality development (6th ed., pp. 505-570). New York: Wiley.

17.

Harter

Pike

(1984). The Pictorial Scale of Perceived Competence and Social Acceptance for Young Children. Child Development, 55, 1969-1982.

18.

Hayward

D. L.

(2004). The emergence of the collective and the individual self: A cross-cultural study of Japanese and American children (Doctoral dissertation, Alliant International University, 2004). Dissertation Abstracts International, 64, 4039.

19.

Hermans

H. J. M.

Kempen

H. J. G.

(1998). Moving cultures: The perilous problems of cultural dichotomies in a globalizing society. American Psychologist, 53, 1111-1120.

20.

Holmes

H. A.

Black

Miller

S. A.

(1996). A cross-task comparison of false belief understanding in a Head Start population. Journal of Experimental Child Psychology, 63, 263-285.

21.

Hughes

(2002). Executive functions and development: Emerging themes. Infant and Child Development, 11, 201-209.

22.

Kobayashi

Glover

G. H.

Temple

(2007). Cultural and linguistic effects on neural bases of “Theory of Mind” in American and Japanese children. Brain Research, 1164, 95-107.

23.

Kwon

Y.- I.

(2002). Western influences in Korean preschool education. International Education Journal, 3, 153-164.

24.

Kwon

Y.- I.

(2003). A comparative analysis of preschool education in Korea and England. Comparative Education, 39, 479-491.

25.

Lee

Super

C. M.

Harkness

(2003). Self-perception of competence in Korean children: Age, sex, and home influences. Asian Journal of Social Psychology, 6, 133-147.

26.

Leichtman

M. D.

Wang

Pillemer

D. B.

(2003). Cultural variations in interdependence and autobiographical memory: Lessons from Korea, China, India, and the United States. In Fivush

Haden

C. A.

(Eds.), Autobiographical memory and the construction of a narrative self: Developmental and cultural perspectives (pp. 73-97). Mahwah, NJ: Erlbaum.

27.

Liang

K.- Y.

Zeger

S. L.

(1986). Longitudinal data analysis using generalized linear models. Biometrika, 73, 13-22.

28.

Liu

Wellman

H. M.

Tardif

Sabbagh

M. A.

(2008). Theory of mind development in Chinese children: A meta-analysis of false belief understanding across cultures and languages. Developmental Psychology, 44, 523-531.

29.

Lucariello

J. M.

(2004). New insights into the functions, development, and origins of theory of mind: The “Functional-Multilinear Socialization” (FMS) Model. In Lucariello

Hudson

Fivush

Bauer

(Eds.), The development of the mediated mind: Sociocultural context and cognitive development (pp. 33-57). Mahwah, NJ: Erlbaum.

30.

Lucariello

J. M.

Le Donne

Durand

Yarnell

(2006). Social and intrapersonal theories of mind: “I interact therefore I am.” In Antonietti

Liverta-Sempio

Marchetti

(Eds.), Theory of mind and language in developmental contexts (pp. 149-171). New York: Springer.

31.

Markus

H. R.

Kitayama

(1991). Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review, 98, 224-253.

32.

Mascolo

(Eds.). (2004). New directions for child and adolescent development: No. 104. Culture and developing selves: Beyond dichotomization. San Francisco: Jossey-Bass.

33.

Mullen

M. K.

(1994). Earliest recollections of childhood: A demographic analysis. Cognition, 52, 55-79.

34.

Mullen

M. K.

(1995). The cultural context of talk about the past: Implications for the development of autobiographical memory. Cognitive Development, 10, 407-419.

35.

Naito

(2003). The relationship between theory of mind and episodic memory: Evidence for the development of autonoetic consciousness. Journal of Experimental Child Psychology, 85, 312-336.

36.

Naito

Koyama

(2006). The development of false-belief understanding in Japanese children: Delay and difference? International Journal of Behavioral Development, 30, 290-304.

37.

Naito

Seki

(2009). The relationship between second-order false belief and display rules reasoning: The integration of cognitive and affective social understanding. Developmental Science, 12, 150-164.

38.

Nguyen

Frye

(1999). Children’s theory of mind: Understanding of desire, belief and emotion with social referents. Social Development, 8, 70-92.

39.

Nisbett

R. E.

(2003). The geography of thought: How Asians and westerners think differently—and why. New York: Free Press.

40.

Lewis

(2008). Korean preschoolers’ advanced inhibitory control and its relation to other executive skills and mental state understanding. Child Development, 79, 80-99.

41.

Ohtsubo

(2007). Japanese children’s development of false belief understanding. Psychological Reports, 100, 87-97.

42.

Oyserman

Coon

H. M.

Kemmelmeier

(2002). Rethinking individualism and collectivism: Evaluation of theoretical assumptions and meta-analyses. Psychological Bulletin, 128, 3-72.

43.

Raeff

(2006). Always separate, always connected: Independence and interdependence in cultural contexts of development. Mahwah, NJ: Erlbaum.

44.

Rai

Mitchell

(2004). Five-year-old children’s difficulty with false belief when the sought entity is a person. Journal of Experimental Child Psychology, 89, 112-126.

45.

Sabbagh

M. A.

Carlson

S. M.

Moses

L. J.

Lee

(2006). The development of executive functioning and theory of mind: A comparison of Chinese and U.S. preschoolers. Psychological Science, 17, 74-81.

46.

Sampson

E. E.

(1988). The debate on individualism: Indigenous psychologies of the individual and their role in personal and social functioning. American Psychologist, 43, 15-22.

47.

Shatz

Diesendruck

Martinez-Beck

Akar

(2003). The influence of language and socioeconomic status on children’s understanding of false belief. Developmental Psychology, 39, 717-729.

48.

Symons

D. K.

Clark

S. E.

(2000). A longitudinal study of mother-child relationships and theory of mind in the preschool period. Social Development, 9, 3-23.

49.

Symons

D. K.

McLaughlin

Moore

Morine

(1997). Integrating relationship constructs and emotional experience into false belief tasks in preschool children. Journal of Experimental Child Psychology, 67, 423-447.

50.

Tardiff

Wellman

H. M.

Cheung

K. M.

(2004). False belief understanding in Cantonese-speaking children. Journal of Child Language, 31, 779-800.

51.

Toyama

(2007). Examining theory-of-mind tasks with Japanese children: The Wellman and Liu tasks. Japanese Journal of Educational Psychology, 55, 359-369.

52.

Vinden

P. G.

(1999). Children’s understanding of mind and emotion: A multi-culture study. Cognition and Emotion, 13, 19-48.

53.

Vinden

P. G.

(2001). Parenting attitudes and children’s understanding of mind: A comparison of Korean- and Anglo-American families. Cognitive Development, 16, 783-809.

54.

Wellman

H. M.

Cross

Watson

(2001). Meta-analysis of theory-of-mind development: The truth about false belief. Child Development, 72, 655-684.

55.

Wellman

H. M.

Fang

Liu

Zhu

Liu

(2006). Scaling of theory-of-mind understandings in Chinese children. Psychological Science, 17, 1075-1081.

56.

Zeger

S. L.

Liang

K.- Y.

(1986). Longitudinal data analysis for discrete and continuous outcomes. Biometrics, 42, 121-130.